Innovative Mass Air Flow Measurement

Mr. Lucas Howard, Mr. Jonathan Hayes,

Mrs. Pat Buford, Dr. Wayne Helmer,

Mechanical Engineering/Electrical Engineering

Arkansas Tech University
Russellville, Arkansas

Abstract

The purpose of this project is to design and fabricate innovative air flow lab equipment using
standard, low-cost components. The major equipment components are automotive mass air flow
(MAF) sensors and an M68HC11EVB evaluation board. Students using the equipment will take
voltage readings from an LED display of the microcontroller based output of the lab system.
The MAF results will also be compared to mass air flow measurements taken from standard flow
rate devices such as an orifice meter, rotameter or other reference flow rate devices. This lab
equipment was designed to be used by the engineering students in courses ranging from the
freshman introductory class through senior level courses.

Introduction

The faculty of the electrical and mechanical engineering departments at Arkansas Tech
University is constantly trying to improve the laboratory experiences of their students. The
students who use the equipment range from the inexperienced first semester to the seasoned
junior and senior engineering students. Sometimes laboratory equipment that satisfies the
educational needs of this diverse group of students is not economical or readily available.

To satisfy the dual requirements of diversity and cost efficiency, two students, one electrical and
one mechanical, were offered special topic engineering credit to design and build this fluid flow
equipment under the supervision of faculty mentors.

MAF Theory

Automotive mass airflow sensors (MAF) are used to calculate the amount of air entering the
engine intake of an electronic fuel injection system. There are two types of automotive MAFs,
digital and variable voltage (analog). Both use a hot-wire principal.

The hot-wire principal uses a resistor (usually a coil of wire) that is heated by a constant voltage.
The coil of wire is placed in an airflow-sampling channel. See Figure 1. The entering the engine
intake cools the wire. The wire is a positive coefficient resistor, as the temperature decreases the
Proceedings of the 2004 American Society for Engineering Education Midwest Section
Conference
resistance decreases. The decrease in resistance allows more current to flow through the resistor.
The induced change in voltage is sent to the computer of the vehicle to adjust the fuel flow
accordingly.

Air
Flow

Figure 1 Basic hot-wire probe.

Mass flow rate is found by m = AV (mass flow rate = density of air x cross-sectional area x
velocity). The area is easily found by measuring the diameter of the entrance of the mass airflow
sensor. The following steps determine the velocity.

First the supplied voltage and resistance of the resister must be known to determine the power
dissipated by the resistor, P. In most cases, 12 volts are supplied. The resistance (R) can be
measured using a digital multi-meter (DMM).

The equation to find P: P = V2/R.

P = q, q is the energy removed by convection in a moving air stream.

q = d l h (Tw-T)

Where:

q - power dissipated

Proceedings of the 2004 American Society for Engineering Education Midwest Section
Conference
d - diameter of wire

l - length of wire

h heat-transfer coefficient

Tw - temperature of wire

T ambient air temperature

The diameter, length, Tw, and T can be measured. The heat-transfer coefficient can be found.
using the following equation,

h = Nud (k/d)

Where:

k - thermal conductivity of the air

d - wire diameter

The Nusselt number, Nud, can be determined by the following equation. The following equation
will work for most conditions including electrically heated wire in airflow.

Nud = (0.683)(Red) 0.466(Prf) 1/3

Where:

Red - Reynolds number

Prf - Prandtl number

The Prandtl number is a function of the fluid temperature (see below). The Reynolds number
can be determined by using the following equation.

Red = (ud)/vf

Where:

u - velocity of the air stream

d - wire diameter

vf - kinematic viscosity of the air at the film temperature conditions

Proceedings of the 2004 American Society for Engineering Education Midwest Section
Conference
The kinematic viscosity can be found if the film temperature is known. The film temperature
can be determined by using the following equation.

Tf = (T-Tw)/2

The velocity of the air stream can be found when all of the previous equations have been
combined, simplified, and solved for u.

After finding the velocity, determining the density of the air, and measuring the entrance
diameter of the MAF, the following equation can be used to find the mass airflow.

m = ((d/2) 2)()(u)

Where:

d - diameter of MAF sensor entrance

- density of air

u - average velocity of air stream

All of the equations used in the MAF theory were found in referecnece [1] and [2].

Design

The apparatus consists of a frame with a blower, frame, air flow ducting, automobile MAF
sensors, and a microcontroller. A light-emitting diode (LED) displays the voltage from the
sensor. Additionally, students will input the displayed voltage to a C program or their
calculators to determine the MAF.

After reviewing different designs, a design consisting of PVC pipe (for the air flow duct) and a
centrifugal blower was chosen because of its simplicity and low cost of materials. The PVC pipe
would channel the airflow and support the measuring devices. The needed flow rate and the
pressure drop induced by the measuring devices and from frictional losses due to the pipe would
determine the size of the blower.

Blower sizing

The original design was made up of three Ford MAFs (mounted in parallel to each other), a flow
straightener, an orifice manometer, and a rotameter connected in series with 7.25 feet of 3-inch
diameter PCV pipe. The MAFs were mounted in parallel so three student groups could take data
very quickly in sequence. The pressure drop of this configuration was calculated to be 44.51
inches SP, a list of the components and corresponding pressure drops can be found in Table 1.
The needed volumetric flow rate was determined to be 250 cfm. A blower that could exceed
these parameters would cost anywhere from $2200 to $4400. Due to funding limitations a lower

Proceedings of the 2004 American Society for Engineering Education Midwest Section
Conference
cost blower would have to be used. In order to use a blower with lower capabilities, the design
of the system had to be changed. The initial orifice manometer and the rotameter (reference flow
measuring devices) had large pressure drops, 15 inches SP and 25 inches SP respectively. It was
decided that these components needed to be removed and replaced by a commercially available
hot wire probe [4]. The new design consisted of three MAFs (mounted parallel to each other)
and a flow straightener connected in series with 6.5 feet of 3 inch diameter PCV pipe. The
pressure drop for the new design was calculated to be 3.97 inches SP, a list of the components
and corresponding pressure drops can be found in Table 2. A blower-motor assembly that could
handle these requirements cost around $320.

Table 1 Original Design Pressure Drop

Component Pressure loss (in. water)
90 deg elbows 0.92
35 deg turn 0.31
Orifice manometer 15
straws 2.22
Rotameter, King
25
Instruments #0231
Entrance loss 0.31
Pipe 0.75
Total 44.51

Table 2 Final Design Pressure Drop

Final Design
Component Pressure loss (in. water)
90 deg elbows 0.46
35 deg turn 0.31
straws 2.22
Entrance loss 0.31
Pipe 0.67
Total 3.97

Frame

A frame was designed to support the system. The frame had to be rigid, easy to maneuver,
support the apparatus at a comfortable height for students to work on. The overall dimensions
of the frame are 6.5 feet long, 23 inches wide, and 35 inches tall from 70 feet of 1.5-inch square
steel tubing. The basic frame was fabricated and additional cross members and supports were
added when necessary. Steel plates were welded to the four bottom corners of the frame. Caster
wheels were attached to each plate.

Pipe
Proceedings of the 2004 American Society for Engineering Education Midwest Section
Conference
The pipe layout is a simple design. The three MAFs (< $30 total from E-bay) are inserted into a
nominal 3 inch PVC double-wye, as seen in Figure 2. The flow straightener was composed of
about 150 straws bound together with utility tape. To prevent the flow straightener from moving
through the system, a wire screen was attached a foot from the end of a 2-inch pipe. The end
with the flow straightener is connected to the double-wye. Rubber connectors connect an 18-
inch length pipe to the 27-inch length of pipe and to a 7-inch length of pipe. Rubber connectors,
allow for easy removal of the 18-inch section if an orifice manometer is to be added in the future.
The 7-inch length of pipe leads to a 90 elbow. The elbow connects the horizontal line to a
vertical wye. A 2-inch ball valve is connected to the single wye and serves as a by-pass valve for
flow control. After the wye, the flow goes through another 90 elbow to a 3-inch to 6-inch
expansion fitting, then into the blower.

Figure 2. MAFs, frame, and blower-motor assembly

Proceedings of the 2004 American Society for Engineering Education Midwest Section
Conference
Blower and wiring installation

The blower-motor assembly is a capacitor start, single phase 115v, 1.5 hp motor with a high
pressure, direct drive, radial blade blower. The wiring schedule consisted of 20 feet of 10-gage
wire, conduit, conduit connectors, a switch, a conduit box, and a 120V plug. The switch and
conduit box were mounted in a back corner near the blower. It was placed in a location where it
would be unobtrusive but still easily accessible. The cord length from the conduit box to the
plug is approximately 15 feet. This allows the frame to be placed in almost any desired location
in a lab room.

Mass Air Flow Sensor and Circuitry

The MAF sensor used in the automobile industry is basically an air flow to voltage conversion
device using a hot wire anemometer. Measuring mass air flow with this sensor is an application
of both electrical and mechanical engineering principles.

Each MAF has a plate covering the entrance. This is to allow each group to test one MAF at a
time. The MAFs have four connections located in a plastic housing. The connectors are labeled
A, B, C, and D. A 12 V voltage source is connected to connection A. The ground of the voltage
source is connected to connection B. Connection D is for the signal voltage. A volt meter
should be connected to connections B and D, this measures the output voltage. Connection C is
not used. See Figure 3 for wiring diagram. Students should not apply a voltage to a MAF when
the air flow cover is in place since this will damage the platinum wire inside of the flow channel
due to the no-flow situation..

Proceedings of the 2004 American Society for Engineering Education Midwest Section
Conference
 A B C D

Flow

Figure 3 - Wiring connection diagram. [3]

Table 3 Mass airflow and corresponding signal voltage

Mass Airflow (kg/hr) Mass Airflow (lb/hr) Vout (V)

16.77 36.89 1.02
19.19 42.22 1.09
21.41 47.10 1.16
22.93 50.45 1.30
26.26 57.77 1.35
56.43 124.15 2.14
65.51 144.12 2.18
75.07 165.15 2.29
79.61 175.14 2.27
109.11 240.04 2.56

Proceedings of the 2004 American Society for Engineering Education Midwest Section
Conference
 500
450
400 y = 11.666e1.1623x
R2 = 0.9931
Mass Airfow (lb/hr)

350
300
250
200
150
100
50
0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Signal Voltage (V)

Figure 4 Graph of typical mass airflow and signal voltage data

The data shown in the Table 3 and Figure 4 were measured and calculated in a typical student lab
session. This process can be repeated for each MAF.

Microcontroller Interface

The MC68HC11 microcontroller is used as a digital voltmeter for the three MAF sensors. The
program uses Port E as an A/D converter. Students input from a keypad which MAF sensor they
are using in the experiment. VDB (from Figure 4) of the selected MAF sensor is the input to the
A/D. The first task in using the A/D converter is to generate a short delay to power up the A/D.
Then the program must wait for the A/D conversion to be completed. The next step is to get the
input and display the results. Unfortunately the data must be converted from binary to binary
coded decimal so that it may be converted once again, this time to ASCII for display. The
flowchart of Figure 5 outlines the process of a conversion of VDB from a single MAF sensor to
output voltage. Students input the voltage displayed to a C program or their calculators to
compute the actual mass air flow. The program uses the equation of the line graphed in Figure 4
for the calculation.

Unlike the theory of the hot-wire anemometer discussed in texts, the MAF sensor used in
automobiles employ complex circuitry as seen in Figure 6. It was an initial intent to have
students build the simple Wheatstone bridge circuit with an amplifier and then compare their
results to those obtained by the MAF sensor circuitry. Since the concept was to allow students
to experience the practicality of the simple circuit, the complexity of the circuitry onboard the
MAF sensor forced abandonment of this aspect of the laboratory.

Proceedings of the 2004 American Society for Engineering Education Midwest Section
Conference
 Monitor Equates Begin
Constant Equates Input Capture 1
Memory Map Subroutine to end
program
A

Begin Program

Load End of Program

Message
Enable pulse accum. Clear pulse
And IC1 interrupt accumulator flag

Display the output Send end

message The voltage message to
Load X with 121
is Short delay to monitor
Set TFLG2 for
power up A/D
pulse accum flag
B

A/D conversion Check pulse Go to Buffalo

With short wait accumulator flag Return from
interrupt

No
Get the input
multiply by Is it
resolution set?

Jump to BIN-BCD
conversion Decrement X
subroutine

Load result and print

answer X=0 ?

Check pulse accumulator

overflow flag
A

No
Yes
Is it
set? B

Figure 5 Flowchart of analog to digital voltage conversion

Proceedings of the 2004 American Society for Engineering Education Midwest Section
Conference
 Figure 6 MAF Sensor Circuitry

Summary and Conclusions

Air flow laboratory equipment can be extremely expensive. The uniqueness of this project is
that a group of students designed and built the inexpensive (less than $1000) laboratory
equipment which is then be used for years by other students in mechanical and electrical
engineering laboratory experiments. It is a tremendous educational experience for both groups
of studentsthe designers and students enrolled in the laboratory. Additionally, the design
portion of this project provides an excellent opportunity for a multi-disciplinary team of both
electrical and mechanical engineering students.

References

[1] Munson, Young, and Okiishi, Fundamentals of Fluid Mechanics, 3rd ed.
Whiley, New York, 2002.

[2] J.P. Holman, Heat Transfer, 9th ed. McGraw-Hill, New York, 2002.

[3] Using a Mass Airflow Sensor,

http://www.gecdsb.on.ca/sub/projects/ps/notemplates/wind.htm

[4] TSI Corporation. Shoreview MN. Model 8386 velocicalc portable air velocity meter.

Biographical Information

Lucas Howard is a mechanical engineering graduate of Arkansas Tech University and an sales engineer at Jack
Tyler Engineering in Little Rock, Arkansas..

Proceedings of the 2004 American Society for Engineering Education Midwest Section
Conference
Jonathan Hayes is a senior electrical engineering student at Arkansas Tech University.

Patricia Buford is an assistant professor in the electrical engineering department at Arkansas Tech University. She
teaches the early electric circuit course as well as the senior level control systems courses. She is a registered
professional engineer in the state of Arkansas.

Wayne Helmer is a professor in the mechanical engineering department at Arkansas Tech University. He teaches
both the freshman Introduction to Engineering Class as well as senior level courses in heat transfer, HVAC, and
thermodynamics courses. Professor Helmer is a registered professional engineer in the state of Illinois.

Proceedings of the 2004 American Society for Engineering Education Midwest Section
Conference